Back to all tools
Free · runs in your browser

Transcript cleaner

Strip the furniture out of a transcript export: timestamps, speaker labels, [inaudible] markers and cue numbers, then rejoin the lines it was hard-wrapped into. Five passes, each switchable, each reporting how many lines it touched.

Transcript passes

Blank-line gaps left behind by a removed cue block are closed automatically. For anything that is not transcript furniture — double spaces, smart quotes, filler words, hard-wrapped prose, ALL CAPS — use the text cleaner, which has a one-click preset for meeting transcripts.

Lines in
0
Lines out
0
Words out
0

Everything here runs on your device. Nothing you paste is uploaded.

What this tool does, and what it does not

Almost nothing that produces a transcript produces prose. A meeting recorder gives you speaker labels and a timestamp every few seconds. A captions file gives you cue numbers and lines hard-wrapped at 42 characters regardless of where the sentence is. A transcriptionist leaves [inaudible] and [crosstalk] where the audio failed. Paste any of it into a document and you spend twenty minutes deleting furniture by hand.

This page removes that furniture and nothing else. It is deliberately narrow: five passes, all of them about transcript and caption markup. General cleanup — double spaces, smart quotes, filler words, blank lines, ALL-CAPS text, prose that was hard-wrapped by an email client rather than a captioner — lives on the text cleaner, which has a one-click “meeting transcript” preset for exactly this material. Many people run the transcript through here first and the result through there second.

The five passes

Strip timestamps

Catches the shapes that show up in real exports: bracketed stamps like [00:01:23] and [00:01:23.456], cue timing lines like 00:00:04,120 --> 00:00:07,900 (both the comma and the dot decimal separator), parenthesised stamps like (12:34), and a bare 1:23 or 00:01:23 opening a line. A line that was nothing but a timestamp is removed rather than left behind as an empty line.

A full hh:mm:ss at the end of a line is also stripped, but a bare mm:ss there is left alone — “the meeting moved to 3:30” is a sentence, not a marker.

Strip speaker labels

Removes SPEAKER 1:, Speaker 2 :, Participant 3:,Interviewer:, all-caps names like JOHN: and DR. PATEL:, title-case full names like John Doe:, and the >> chevrons that broadcast captions use for a change of speaker. A single title-case word followed by a colon is left alone on purpose, so “Note: the numbers are provisional” survives.

An all-caps clause that merely ends in a colon is protected too: “THE PLAN IS SIMPLE: WE SHIP FRIDAY” comes through intact, because a speaker label never contains words like the, is or we. If you need to keep track of who said what, leave this pass off and delete the labels by hand at the end.

Strip [inaudible] markers

The bracketed asides a transcriptionist leaves in place of audio: [inaudible], [indiscernible], [crosstalk], [laughter], [applause], [pause], [BLANK_AUDIO], [music], [phonetic], (unintelligible) and the timestamped variants like [inaudible 00:14:02]. Square brackets, round brackets and braces are all handled. Brackets that contain anything else — [the shared folder], a citation, an editorial insertion — are left exactly where they are.

Removing these changes the meaning of the transcript: a sentence that reads “the number was [inaudible] million” becomes “the number was million”. That is usually what you want when the transcript is becoming an article, and never what you want when it is evidence. Read the result before you use it.

Strip line and cue numbers

Off by default, because a numbered list looks exactly like a numbered transcript to a regular expression. It removes a bare number on its own line — the cue index in an SRT file — and the 12. / 12) / 12: prefixes used by line-numbered transcripts. Switch it on for a captions or deposition export; leave it off for anything with an ordered list in it.

Unwrap hard-wrapped lines

Joins a line onto the one above it when that line ended mid-sentence — that is, when it did not end in ., !, ?, :, ; or a closing quote or bracket. A blank line always starts a new paragraph, a line beginning with a bullet or a list number is never swallowed into the one above, and a line that is nothing but a number — a cue index you chose to keep — never has the sentence below it pulled up alongside it.

This is the pass that turns a 42-character caption column back into paragraphs. It is conservative on purpose: a transcript with one sentence per line keeps its structure, because each of those lines already ends in a full stop.

The order the passes run in

The order is fixed and it matters. Cue numbers come off first, then timestamps, then speaker labels, then markers — a sentence-level rule cannot reason about a line that starts 00:04:12 SPEAKER 2:. Unwrapping runs last, because it has to judge where a sentence ended, and it can only do that once the furniture is gone. The blank-line gaps left where a cue block used to be are closed automatically at the end; that is not a decision worth a checkbox.

One caveat worth knowing: unwrapping and speaker labels do not always sit well together. If a turn is cut off mid-sentence by a hard wrap, the next speaker’s line can be joined onto it. Clean a labelled transcript with unwrapping off, or strip the labels and let the paragraphs reflow.

Where the mess comes from — and how to stop making it

Transcript furniture exists because the recording came first and the writing came second. Every meeting tool has to tell you when and who, so it interleaves that with the words, and you spend your time undoing it.

VoiceSnap Pro is a voice-to-text dictation and voice-notes app for macOS and Windows. Hold one keyboard shortcut, speak, and punctuated text appears in whatever field your cursor is already in: an email, a ticket, a document, a chat. Your personal vocabulary helps the app recognize names and jargon, and a second shortcut saves what you say as a voice note in your account, where you can tag it, file it in a folder, rework it with AI actions or have it remind you later. Nothing you dictate is used to train AI models.

Dictation inverts the order: the text is written straight into the document you are working on, so there is no export step to clean up.

Start with 5,000 transcribed words across dictation and voice notes, plus 50 manual AI actions, given once with no expiry, every feature and no card. Pro is $9 a month, subject to fair use. Download free for Mac or Windows. The tool on this page stays free either way, with no account.

Questions people ask

Further readingCommonplace Book You Can SearchA longer read on the VoiceSnap Pro blog.